Oxide thermistor compositions

Abstract

Oxide thermistor compositions which comprise 100 atomic % of at least four kinds of cations which are (1) Mn ion, (2) Ni ion, (3) at least one kind of ion selected from the group consisting of Cu, Fe, and Cr, and (4) one kind of ion selected from the group consisting of Cr, Zr, and Li. These compositions have lower resistivity with higher B-constant and exhibit a high stability of resistance.

Description

The present invention relates to oxide compositions for thermistors.

Thermistors containing primarily Mn-oxide and additionally Co-oxide have been widely used until now. The reasons why the thermistors of Co-oxide-containing composition have been widely used are due to the excellent thermistor properties thereof such as (1) higher B-constant (which can be obtained) together with low resistivity and (2) a smaller resistance (duration in load aging in the temperature below 300.degree. C. under an application of a d.c. voltage.) Thermistor materials having decreased resistivity have as a rule decreased B-constant. Accordingly, it can be said that a material having a low resistivity together with a higher B-constant is useful as a thermistor.

However, Co-oxide sources have recently become difficult to obtain and more expensive throughout the world, and this has developed a need for a thermistor composition containing no Co-oxide, which also exhibits excellent thermistor properties comparable to those of Co-oxide-containing thermistor compositions.

An object of the present invention is to provide oxide thermistor compositions containing no Co-oxide.

Another object of the present invention is to provide oxide compositions for thermistors having high stable electrical characteristics in load aging under an application of a d.c. voltage.

A further object of the present invention is to provide oxide compositions for thermistors having lower resistivity with higher B-constant.

The oxide thermistor compositions of the present invention are characterized by containing primarily Mn-oxide and additionally Ni-oxide, at least one kind of oxide selected from the group consisting of Cu-oxide, Fe-oxide, and Zr-oxide, and one kind of oxide selected from the group consisting of Cr-oxide, Zr-oxide, and Li-oxide.

The effect of Cr-oxide contained in the compositions of the present invention is to provide a high stability of resistivity; the effect of Zr-oxide therein is to provide a relative stability of resistivity and also relatively high B-constant; and the effect of Li-oxide therein is to provide a B-constant relatively high for the resulting low resistivity.

Referring to the prior art of the thermistor compositions which contain primarily Mn-oxide and additionally Cr-oxide, only the following systems have been disclosed:

Mn-Cr oxide systems (Hitachi Central Lab. Tech. Papers, the Memorial Edition for the 20th Anniversary of the Establishment, 1962).

Mn-Ni-Cr oxide systems [Denki Kagaku, Vol. 19, No. 9, (1951)] ##EQU1##

Type of oxide thermistor compositions of the present invention includes

Mn-Ni-Cu-Cr oxides,

Mn-Ni-Fe-Cr oxides, and

Mn-Ni-Cu-Fe-Cr oxides.

These compositions are based upon the finding of the fact, as an effect of the contained chromium which is a feature of the present compositions, that the percentage of resistance deviation thereof in the lapse of 3000 hours in load aging under an application of a d.c. voltage of 10 V/mm at the temperature of 150.degree. C. is as small as .+-.2%, in other words, upon the finding that Cr-oxide has such an effect to stabilize electrical characteristics of thermistors.

Referring to the prior art of thermistor compositions which contain primarily Mn-oxide and additionally Zr-oxide, only one example, i.e., Mn-Zr oxide systems (Hitachi Central Lab. Tech. Papers, the Memorial Edition for the 20th Anniversary of the Establishment, 1962) has been disclosed.

Type of oxide thermistor compositions of the present invention includes

Mn-Ni-Cu-Zr oxides,

Mn-Ni-Fe-Zr oxides,

Mn-Ni-Cr-Zr oxides, and

Mn-Ni-Fe-Cr-Zr oxides.

These compositions are based upon the finding of an effect of the contained Zr, which is a feature of the this type of the composition of the present invention, giving relatively stable electrical characteristics and a B-constant relatively high for the resulting low resistivity.

Referring to the prior art of the thermistor compositions which contain primarily Mn-oxide and additionally Li-oxide, only the following systems have been disclosed:

Mn-Li oxide, Mn-Ni-Li oxide, Mn-Cu-Li oxide and Mn-Fe-Li oxide systems (Hitachi Central Lab. Tech. Papers, the Memorial Edition for the 20th Anniversary of the Establishment, 1962).

Type of oxide thermistor compositions of the present invention includes

Mn-Ni-Cu-Li oxides,

Mn-Ni-Fe-Li oxides, and

Mn-Ni-Cu-Fe-Li oxides.

These compositions are based upon the finding of an effect of the contained Li, which is a feature of this type of the composition of the present invention, giving a B-constant relatively high for the resulting low resistivity.

DETAILED DESCRIPTION OF THE INVENTION

The thermistor compositions of the present invention which are characterized by containing chromium comprise as cations 94.6-30 atomic % of Mn ion, 5-30 atomic % of Ni ion, 0.1-15 atomic % of Cu ion, and 0.3-40 atomic % of Cr ion, the total amount of said cations being 100 atomic %. In this place, a Cr content less than 0.3 atomic % has no observable high stability of resistivity in load aging at the temperature of 150.degree. C. under an application of a d.c. voltage. The Cr content range wherein this effect is remarkable is from 3 to 30 atomic %. A Cr content exceeding 40 atomic % gives a high resistivity coupled with a high B-constant, which is undesirable because it departs from the range of the electrical characteristic values required for practical use. The reason for limiting each content of Mn, Ni, and Cu is based on the electrical characteristic values of the existing general purpose NTC thermistors commercially available, that is to say, the limitation is intended to secure a practical resistivity at 25.degree. C. staying within the range of 10 .OMEGA.cm to 1 M .OMEGA.cm and also a B-constant staying within the range of 1000.degree. K. to 6000.degree. K. With electrical characteristic values out of these ranges, the compositions are deficient in practical usefulness. The resistivity of a thermistor of this type at 25.degree. C. (.rho.25.degree. C.) decreases with an increase in Ni-to-Mn ratio, reaching a minimum at a Ni content of 22 atomic %, and then over this point it conversely begins to rise with the Ni content. On the other hand, the B-constant only decreases a little with an increase in the Ni content, exhibiting a somewhat vague peak at a Ni content of 17.5 atomic % (corresponding to phase transition). In addition, the .rho.25.degree. C. and B-constant both decrease when the Cu content is raised versus the Mn content. As a result, the .rho.25.degree. C. of a composition having a Ni content smaller than 5 atomic % with a Cu content smaller than 0.1 atomic % is out of the range of resistivity acceptable for practical use. Moreover, a Ni content over 30 atomic % is undesirable for thermistor materials because it gives an increased .rho.25.degree. C. together with a decreased B-constant. A Cu content over 15 atomic % is also undesirable for practical thermistors because it gives the markedly decreased values of both .rho.25.degree. C. and B-constant.

Secondly, referring to the Mn-Ni-Fe-Cr oxide and Mn-Ni-Cu-Fe-Cr oxide compositions, there is an observable high stability of resistivity in the load aging similar to that in the above Mn-Ni-Cu-Cr oxide compositions. However, the .rho.25.degree. C. is rather low in comparison with the case of the Mn-Ni-Cu-Cr oxide compositions.

The thermistor compositions of the present invention, which are characterized by containing zirconium, comprise as cations 94.6-50 atomic % of Mn ion, 5-25 atomic % of Ni ion, 0.1-5 atomic % of Cu ion, and 0.3-20 atomic % of Zr ion, the total amount of said cations being 100 atomic %. In this place, a Zr content smaller than 0.3 atomic % has no observable effect of giving a B-constant relatively high for the resulting low resistivity. The Zr content range wherein this effect is remarkable is from 0.5 to 10 atomic %. A Zr content over 15 atomic % results in electrical characteristics of a B-constant relatively low for the resulting high resistivity. With a Ni content smaller than 5 atomic % together with a Cu content smaller than 0.5 atomic %, the .rho.25.degree. C. is much higher, departing from the range of resistivity appreciable for practical use. A Ni content over 25 atomic % is also undesirable because it gives an increased .rho.25.degree. C. value and in addition a decreased B-constant. Further, a Cu content over 5 atomic % is undesirable for thermistors for practical use, because it markedly reduces both .rho.25.degree. C. and B-constant.

The further thermistor compositions of the present invention which are characterized by containing zirconium comprise as cations 94.6-50 atomic % of Mn ion, 5-25 atomic % of Ni ion, 0.15-5 atomic % of Cu ion, and 0.3-20 atomic % of Zr ion, the total amount of said cations being 100 atomic %. In this place, a Zr content smaller than 0.3 atomic % has no observable effect of giving a B-constant relatively high for the resulting low resistivity. A Zr content over 10 atomic % gives characteristics of high resistivity with high B-constant, which is undesirable because of departing from the range of the electrical characteristic values required for practical use. A total content of Fe and/or Cr of smaller than 0.3 atomic % has no high stability of resistivity in load aging at the temperature of 150.degree. C. under an application of a d.c. voltage. A total content of Fe and/or Cr of larger than 5 atomic % is undesirable because it gives a high resistivity which is out of the range of the characteristic values required for practical use. More unfavorably, such a content reduces the sintering capability.

The thermistor compositions of the present invention which are characterized by containing lithium comprise as cations 94.8-50 atomic % of Mn ion, 5-25 atomic % of Ni ion, 0.1-5 atomic % of Cu ion, and 0.1-20 atomic % of Li ion, the total content of said cations being 100 atomic %. In this place, a Li content smaller than 0.1 atomic % has no effect of giving the characteristics of a B-constant relatively high for the resulting low resistivity. The Li content range wherein this effect is remarkable is from 1 to 15 atomic %. A Li content over 20 atomic % results in characteristics of high resistivity with high B-constant, in other words, this is undesirable for the purpose of the present thermistor compositions since only the resistivity shows an increased value while the B-constant shows practically no increased value. With a Ni content smaller than 5 atomic % together with a Cu content smaller than 0.1 atomic %, the .rho.25.degree. C. is much higher, departing from the range of the proper resistivity for practical use. A Ni content over 25 atomic % is also undesirable for the purpose of the present thermistor compositions, because it gives an increased resistivity with a decreased B-constant. With a Cu content over 5 atomic %, it gives markedly decreased values of both .rho.25.degree. C. and B-constant, which are undesirable as characteristics of thermistors for practical use.

In the Mn-Ni-Fe-Li oxide compositions, there is also observed the effect of the added lithium, i.e., characteristics featured by a B-constant relatively high for the low resistivity. An only difference from the Mn-Ni-Cu-Li oxide compositions is that the level of the .rho.25.degree. C. of these compositions is about one order higher than that of the above compositions. In the Mn-Ni-Cu-Fe-Li oxide compositions, however, the .rho.25.degree. C. is observed to be rather small as compared with the case of the Mn-Ni-Cu-Li oxide compositions, when the total content of Cu and Fe does not exceed 5 atomic %. In these Mn-Ni-Cu-Fe-Li oxide compositions, there is equally observed the effect of the added lithium, i.e., the characteristics featured by a B-constant relatively high for the resulting low resistivity.

The commercial powdered compounds, MnCO.sub.3, NiO, CuO, Fe.sub.2 O.sub.3, Cr.sub.2 O.sub.3, ZrO.sub.2, and Li.sub.2 CO.sub.3, were blended as the raw materials to give each of the compositions represented by atomic % in Tables 1, 2, and 3. To illustrate the process for preparing thermistors, the blended composition was wet mixed in a ball mill; the resulting slurry was dried, and then calcined at 800.degree. C.; the calcined material was wet mixed and ground in a ball mill; the resulting slurry was dried and polyvinyl alcohol was admixed therewith as a binder; therefrom a number of the required amount of the mass were taken and each was pressed to form a disk; these disks were sintered in the air at 1100.degree. C. (the sintering temperature for producing practical thermistors can be varied within the range of 1000.degree.-1200.degree. C.) for 2 hours; each of two electrodes comprising silver as main constituent was baked on each side surface of the sintered disk (about 7 mm in diameter and 1.5 mm in thickness) to obtain ohmic contact. The resistance was measured on these specimens at 25.degree. and 50.degree. C. (R.sub.25.degree. C. and R.sub.50.degree. C.), and therefrom the resistivity at 25.degree. C. (.rho.25.degree. C.) and the B-constant were calculated using the following formulae (1) and (2), respectively: ##EQU2## (S: surface area of either of the electrodes; d: distance between the two electrodes) ##EQU3##

In order to evaluate the stability of resistance of each specimen, a d.c. voltage of 10 V/mm was applied to each specimen in a thermostat of 150.degree. C. to measure the resistance deviation with time during 3000 hours. These results were shown in Tables, 1, 2, and 3.

TABLE 1 __________________________________________________________________________ Percentage Composition of sample (atomic % of resis- Sample of constituent) .rho.25.degree. C. B tance with No. Mn Ni Cu Fe Cr [.OMEGA. . cm] [K] time (%) __________________________________________________________________________ 101* 80 17.5 2.5 0 0 820 3900 +3.8 102* 79.9 17.5 2.5 0 0.1 800 3900 +3.3 103 79.8 17.4 2.5 0 0.03 800 3900 +1.8 104 79.2 17.3 2.5 0 1.0 870 3950 +1.3 105 77.6 17.0 2.4 0 3.0 900 4000 +0.7 106 72.0 15.8 2.2 0 10 1250 4100 .+-.0.2 107 64.0 14.0 2.0 0 20 2400 4300 .+-.0.2 108 48.0 10.5 1.5 0 40 8.3 .times. 10.sup.4 5300 .+-.0.2 109* 40.0 8.8 1.2 0 50 2.1 .times. 10.sup.4 6200 -0.5 121* 94.9 5.0 0.1 0 0 1.4 .times. 10.sup.4 4800 +3.5 122 94.6 5.0 0.1 0 0.3 9.0 .times. 10.sup.4 4800 +1.7 123* 50 33 17 0 0 4.5 1300 +2.2 124 45 30 15 0 10 120 2900 +0.8 124 30 20 10 0 40 2.0 .times. 10.sup.4 3900 +0.5 125* 20 35 5 0 40 1.3 .times. 10.sup.4 5400 .+-.0.2 201* 80 17.5 0 2.5 0 3200 4000 +4.5 202 79.8 17.4 0 2.5 0.3 3100 4000 +1.9 203 77.6 17.0 0 2.4 3 5300 4100 +1.3 204 72.0 15.8 0 2.2 10 1.1 .times. 10.sup.4 4400 +1.0 205 48.0 10.5 0 1.5 40 6.4 .times. 10.sup.5 4900 +0.9 206* 40.0 8.8 0 1.2 50 7.3 .times. 10.sup.4 5400 +1.0 211 65.0 20.0 0 5.0 10 2.3 .times. 10.sup.4 4300 +1.2 212 55 25 0 10 10 4.1 .times. 10.sup.5 4700 +1.1 213* 20 40 0 20 20 1.5 .times. 10.sup.4 5100 +1.3 301* 74 20 5.0 1.0 0 130 3100 +3.9 302 73.8 19.9 5.0 1.0 0.3 120 3100 +1.8 303 66.6 18.0 4.5 0.9 10 630 3500 .+-.0.2 304 44.4 12.0 3.0 0.6 40 5.4 .times. 10.sup.4 4800 .+-.0.2 305* 37.0 10.0 2.5 0.5 50 1.4 .times. 10.sup.4 6000 .+-.0.2 311 62.0 20.0 7.0 1.0 10 400 3300 +0.5 312 60.0 20.0 7.0 3.0 10 350 3100 +0.9 313 56.0 20.0 7.0 7.0 10 210 3000 +1.8 __________________________________________________________________________ *The star mark represents a referential sample for comparison, which is out of the scope of the present invention.

Samples 109, 121, 125, 206, 213,and 305 have exhibited .rho.25.degree. C. values in excess of 1 M .OMEGA.cm and therefore are deficient in practical usefulness, departing from the scope of the present invention. Sample 123 has a .rho.25.degree. C. value lower than 10 .OMEGA.cm, which lies out of the range of proper resistivity for practical use. Samples 101, 102, 121, 123, 201, and 301 were regarded as being out of the scope of this invention because there was no indication of receiving the effect of the added chromium, which is an object of this invention, i.e., the objective effect is that the percentage of the resistivity deviation after 3000 hours' load aging under the above-mentioned conditions is not more than .+-.2%. All the samples that are within the scope of this invention have thermistor properties lying within the range of the electrical characteristic values required for practical use, and on all these samples the effect of the added chromium, i.e., resistance-stabilizing effect has been observed. This indicates that these samples can be put to practical use with satisfaction.

In the preparation of the above samples, agate balls were used for mixing the raw materials and for mixing and grinding the calcined materials. The results of elementary analysis on the above samples (sintered mass) showed that in every sample the total content of the contaminating, glass forming elements such as silicon and boron was not more than 1 atomic % per 100 atomic % of the thermistor constituting elements. Subsequently, the composition of sample 106 was selected out, blended with powdered silica to give Si contents of 1 and 2 atomic % per 100 atomic % of the thermistor-constituting elements, and processed in the same way and under the same conditions as used in preparing the above samples, to prepare two kinds of thermistor samples. As a result, the thermistor containing 1 atomic % of Si showed a .rho.25.degree. C. value of 1320 .OMEGA.cm, a B-constant of 4100.degree. K., and a percentage of the abovementioned time-dependent resistance deviation of +0.5%, which are almost the same as those of sample 106, whereas the thermistor containing 2 atomic % of Si showed a .rho.25.degree. C. value of 2700 .OMEGA.cm, a B-constant of 4200.degree. K., and a percentage of the time-dependent resistance deviation of +1.2%. The latter sample, in comparison with sample 106, has a .rho.25.degree. C. much higher (roughly twice) and a higher percentage of the time-dependent resistance deviation, which are undesirable for the objective thermistors of the present invention.

As mentioned above, this invention provides highly stable thermistor compositions, exhibiting extremely small percentages of the resistance deviation in load aging at the temperature of 150.degree. C. under an application of a d.c. voltage.

TABLE 2 ______________________________________ Composition of sample Sample (atomic %) .rho.25.degree. C. B No. Mn Ni Cu Zr (.OMEGA. cm) (.degree.K) ______________________________________ 1101* 80 17.5 2.5 0 820 3900 1151 79.8 17.5 2.5 0.3 825 3880 1152 79.5 17.5 2.5 0.5 810 3940 1153 77.3 17.2 2.5 3.0 822 3970 1154 72.0 15.5 2.5 10.0 840 4030 1155 68.0 14.5 2.5 15.0 1880 3680 1156 65.0 12.5 2.5 20.0 6370 3660 1157 63.0 20.2 6.8 10.0 164 5230 1158 79.7 17.2 0.1 3.0 1940 3850 1159 94.6 5 0.1 0.3 3.2 .times. 10.sup.5 4380 1160 55.0 25.0 10.0 10.0 87.4 3070 1401* 83.3 0 13.7 3.0 411 3130 1501* 79.5 17.5 0 3.0 2320 3870 ______________________________________ (A star mark represents a referential sample for comparison, which is out of the scope of the present invention.)

Samples 1101, 1401, and 1501 are of ternary system and have resistances all lying within the value range acceptable for practical use. However, as can be seen from Table 3, these samples do not satisfy the requirements for the objective thermistors of the present invention, i.e., the requirements including relatively low resistance, relatively high B-constant, and in addition a smaller dependence of resistivity on the sintering temperature. Consequently, these have been regarded as being out of the scope of the present invention. Sample 1101 has obviously a composition of the prior art.

TABLE 3 ______________________________________ Sintering Sam- Composition of sample Resistivity temper- ple (atomic %) .rho.25.degree. C. B-constant ature No. Mn Ni Cu Zr (.OMEGA. cm) (.degree.K) (.degree.C.) ______________________________________ 591 3750 1150 1101* 80 17.5 2.5 0 820 3900 1100 1090 3870 1050 630 3850 1150 1151 79.8 17.5 2.5 0.3 825 3880 1100 1110 3900 1050 674 3980 1150 1154 77.3 15.5 2.5 3.0 840 4030 1100 1170 4050 1050 297 3000 1150 1401* 83.3 0 13.7 3.0 411 3130 1100 466 3140 1050 2320 3800 1150 1501* 79.5 17.5 0 3.0 2840 3920 1100 4130 3940 1050 ______________________________________

All the samples included within the scope of the present invention have properties lying within the range of characteristic values required for practical use. They show the characteristics of low resistance coupled with high B-constant which are the effects brought about by the addition of zirconium and through the adjustment of resistivity by the addition of copper. The percentages of resistance deviation thereof after 1000 hours' continuous load aging in the high humidity (95% RH at 40.degree. C.) under an application of a d.c. voltage (10 V/mm) are within the range of .+-.5%, and those after 3000 hours' continuous load aging at the temperature of 150.degree. C. in the air under an application of a d.c. voltage (10 V/mm) are also within the range of .+-.5%. This indicates that these samples can be put to practical use with satisfaction.

In the preparation of the above samples, agate balls were used for mixing the raw materials and for mixing and grinding the calcined materials. The results of elementary analysis on the above samples (sintered mass) showed that in every sample the total content of the contaminating, glass forming elements such as silicon and boron was not more than 1 atomic % per 100 atomic % of the thermistor-constituting elements. Subsequently, the composition of sample 1154 was selected out, blended with powdered silica to give Si contents of 1 and 2 atomic % per 100 atomic % of the thermistor-constituting elements, and processed in the same way and under the same conditions as used in preparing the above samples, to prepare two kinds of thermistor samples. As a result, the thermistor containing 1 atomic % of Si showed a .rho.25.degree. C. value of 852 .OMEGA.cm and a B-constant of 4040.degree. K., which are almost the same as those of sample 1154, whereas the thermistor containing 2 atomic % of Si showd a .rho.25.degree. C. value of 1500 .OMEGA.cm and a B-constant of 4050.degree. K. In the latter sample, only the .rho.25.degree. C. is much higher (roughly twice) in comparison with sample 1154, which is undesirable for the objective thermistors of the present invention.

TABLE 4 __________________________________________________________________________ Composition of sample Percentage of Sample (atomic %) .rho. 25.degree. C. B resistance devia- No. Mn Ni Fe Cr Zr (.OMEGA.cm) (.degree.K.) tion with time (%) __________________________________________________________________________ 2001* 80 17.5 2.5 0 0 3200 4000 +4.5 2002* 80 17.5 0 2.5 0 2950 4000 +2.1 2003* 80 17.5 0 0 2.5 2200 3840 +4.8 2004 82.4 17.0 0.3 0 0.3 3400 4090 +1.8 2005 75.5 16.0 2.5 0 5 3500 4000 .+-.1.2 2006 94.4 5.0 0.3 0 0.3 9.6 .times. 10.sup.5 4800 +2.0 2007 55 30 0 5 10 7200 4120 +1.9 2008 69.7 20 0 0.3 10 2800 3890 +0.8 2009 86.0 10.0 0 4.0 10 4.7 .times. 10.sup.4 4430 +1.7 2010 85.7 8.0 0.3 1.0 5 2.3 .times. 10.sup.4 4200 +2.1 2011 64 16 5 5 10 3.4 .times. 10.sup.5 4570 +2.0 2012 81.2 17.5 0.5 0.5 0.3 2300 4020 +0.8 __________________________________________________________________________ (A star mark represents a referential sample for comparison, which is out of the scope of the present invention.)

Samples 2001, 2002, and 2003, which are shown for comparison, have large percentages of the time-dependent resistance deviation, lacking in the stability necessary for practical use. Samples 2004 to 2012 showed a high stability, which is an object of the present invention, due to the effect of Fe or Cr and of Zr, i.e., the percentages of resistance deviation thereof after 3000 hours' load aging under the above-mentioned conditions were within the range of .+-.2%. In addition, these samples have properties lying within the range of electrical characteristic values required for practical use. Thus, these samples can be put to practical use with satisfaction.

TABLE 5 ______________________________________ Composition of sample Sample (atomic %) .rho.25.degree. C. B No. Mn Ni Cu Fe Li (.OMEGA. . cm) [.degree.K] ______________________________________ 101* 80 17.5 2.5 0 0 820 3900 102* 79.97 17.5 2.5 0 0.03 810 3900 103 79.9 17.5 2.5 0 0.1 800 3950 104 79.8 17.4 2.5 0 0.3 780 4000 105 79.2 17.3 2.5 0 1.0 750 4050 106 77.6 17.0 2.4 0 3.0 700 4150 107 72.0 15.8 2.2 0 10.0 710 4300 108 68.0 14.9 2.1 0 15.0 780 4550 109 64.0 14.0 2.0 0 20.0 890 4700 110* 60.0 13.1 1.9 0 25.0 1130 4700 121* 94.3 5.6 0.1 0 0 1.2 .times. 10.sup.6 4650 122 84.9 5.0 0.1 0 10 51000 4700 123* 66.7 27.8 5.5 0 0 4.5 1500 124 60 25.0 5.0 0 10 120 3850 125 50 25.0 5.0 0 20 38 2900 126 94.8 5.0 0.1 0 0.1 8.0 .times. 10.sup.5 4700 201* 80 17.5 0 2.5 0 3200 4000 202 79.9 17.5 0 2.5 0.1 3000 4050 203 79.2 17.3 0 2.5 1.0 2900 4250 204 72.0 15.8 0 2.2 10.0 2600 4600 205 64.0 14.0 0 2.0 20.0 3500 4750 206* 60.0 13.1 0 1.9 25.0 4400 4750 211* 94.9 5.0 0 0.1 0 1.4 .times. 10.sup.6 4800 212 94.8 5.0 0 0.1 0.1 9.2 .times. 10.sup.5 5000 213 50.0 25.0 0 5.0 20 41 3000 214* 44.0 30.0 0 6.0 20 5.1 1600 301* 78.5 17.5 3.0 1.0 0 550 3600 302 78.4 17.5 3.0 1.0 0.1 510 3650 303 77.7 17.3 3.0 1.0 1.0 480 3750 304 70.6 15.8 2.7 0.9 10.0 430 4000 305 62.8 14.0 2.4 0.8 20 590 4350 306* 58.9 13.1 2.3 0.8 25 730 4300 311 71.6 15.8 2.2 0.4 10.0 630 4250 312 71.2 15.8 2.2 0.8 10.0 480 4200 313 70.4 15.8 2.2 1.6 10.0 500 4250 314 69.6 15.8 2.2 2.4 10.0 620 4250 ______________________________________ (The star mark represents a referential sample for comparison, which is out of the scope of the present invention.)

Samples 3121 and 3211 showed .rho.25.degree. C. values not smaller than 1 M .OMEGA..cm, being out of the range of the practically appreciable values. Samples 3123 and 3214 showed .rho.25.degree. C. values not larger than 10 .OMEGA.cm, being also out of the range of the practically appreciable values. These have obviously compositions of the prior art. Samples 3101, 3201, and 3301, though exhibiting practically useful resistivity values, have compositions of the prior art. Samples 3110, 3206, and 3306, though having practically useful resistivity values, showed no effect given by the added Li, i.e. the low resistivity coupled with high B-constant characteristics, which are intended by the present invention, and these samples, wherein Li content is over 20 atomic %, are inferior in a stability of resistivity in load aging at a high humidity under an application of a d.c. voltage. From these respects, these samples have been regarded as being out of the scope of the present invention. Showing no effect given by the added Li, sample 3102 has also been regarded as being out of the scope. Meanwhile, the samples of the present invention all have properties lying within the range of practically appreciable characteristic values. They showed the effect given by the added Li and the effect of giving the characteristics of low resistivity coupled with high B-constant. The percentages of resistance deviation thereof after 3000 hours' continuous load aging at the high humidity (95% RH at 40.degree. C.) under an application of a d.c. voltage (10 V/mm) were within .+-.5%, and the percentages of resistance deviation thereof after 3000 hours' continuous load aging at the temperature of 150.degree. C. in the air under an application of a d.c. voltage (10 V/mm) were also within .+-.5%. Consequently, these samples can be put to practical use with satisfaction.

In the preparation of the above samples, agate balls were used for mixing the raw materials and for mixing and grinding the calcined materials. The results of elementary analysis on the above samples (sintered mass) showed that in every sample the total content of the contaminating, glass forming elements such as silicon and boron was not more than 1 atomic % per 100 atomic % of the thermistor-constituting elements. Subsequently, the composition of sample 3107 was selected out, blended with powdered silica to give Si contents of 1 and 2 atomic % per 100 atomic % of the thermistor-constituting elements, and processed in the same way and under the same conditions as used in preparing the above samples, to prepare two kinds of thermistor samples. As a result, the thermistor containing 1 atomic % of Si showed a .rho.25.degree. C. value of 730 .OMEGA.cm and a B-constant of 4300.degree. K., which are almost the same as those of sample 3107, whereas the thermistor containing 2 atomic % of Si showed a .rho.25.degree. C. value of 1500 .OMEGA.cm and a B-constant of 4350.degree. K. In the latter sample, in comparison with sample 3107, the .rho.25.degree. C. is much higher (roughly twice) for the value of B-constant, which is undesirable for the objective thermistor of this invention.

As can be seen from the foregoing description, this invention can provide oxide thermistor compositions of low resistance coupled with high B-constant.

Claims

1. Oxide thermistor composition consisting essentially of metal oxide, wherein said metal consists essentially of 94.6 to 30 atomic % of Mn ion, 5 to 30 atomic % of Ni ion, 0.1 to 15 atomic % of at least one kind of ion selected from the group consisting of Cu and Fe, and 0.3 to 40 atomic % of Cr ion.

2. Oxide thermistor composition consisting essentially of metal oxide, wherein said metal consists essentially of 94.8 to 50 atomic % of Mn ion, 5 to 25 atomic % of Ni ion, 0.1 to 5 atomic % of at least one kind of ion selected from the group consisting of Cu and Fe, and 0.1 to 20 atomic % of Li ion.

3. Oxide thermistor composition consisting essentially of metal oxide, wherein said metal consists essentially of 94.6 to 55 atomic % of Mn ion, 5 to 25 atomic % of Ni ion, 0.1 to 10 atomic % of Cu ion, and 0.3 to 10 atomic % of Zr ion.

4. Oxide thermistor composition consisting essentially of metal oxide, wherein said metal consists essentially of 94.4 to 55 atomic % of Mn ion, 5 to 30 atomic % of Ni ion, 0.3 to 5 atomic % of at least one kind of ion selected from the group consisting of Fe and Cr, and 0.3 to 10 atomic % of Zr ion.

5. Oxide thermistor composition as claimed in claim 1, 2, 3 or 4, which further consists essentially of a binder of less than 1 atomic % of at least one kind of cation selected from the group consisting of Si and B.